docs: drop retired full-graph settings and model keys

content/docs/build/knowledge.mdx

@@ -149,15 +149,6 @@ cuga start demo_knowledge --knowledge-profile max_quality
 
 You can also override embeddings from the CLI (`--embeddings-provider`, `--embeddings-model`, `--embeddings-base-url`, …). Built-in embedding families: local `fastembed` (default), Hugging Face, OpenAI-compatible, Ollama, and OpenRouter.
 
-## Routing Knowledge Through CugaLite
-
-The `[advanced_features].force_lite_mode_apps` list defaults to `["knowledge"]`, so knowledge queries always run through CugaLite's faster execution path regardless of `lite_mode_tool_threshold`. To change this, edit `settings.toml`:
-
-```toml
-[advanced_features]
-force_lite_mode_apps = ["knowledge", "crm"]   # add more apps as needed
-```
-
 <Callout type="info">
 The agent's awareness block is rebuilt as documents are ingested or removed, so newly added documents are usable immediately on the next invocation.
 </Callout>

content/docs/build/memory.mdx

@@ -30,14 +30,10 @@ If Memory is unavailable, times out, or returns nothing, CUGA continues — it n
 Memory is used in **CugaLite**. Turn it on in `settings.toml`:
 
 ```toml
-[advanced_features]
-lite_mode = true
-
 [evolve]
 enabled = true
 mode = "auto"
 app_name = "evolve"
-lite_mode_only = true
 save_on_success = true
 save_on_failure = true
 async_save = true
@@ -66,10 +62,10 @@ EVOLVE_BACKEND=postgres
 EVOLVE_PG_HOST=localhost
 EVOLVE_PG_PORT=5432
 EVOLVE_PG_USER=postgres
-EVOLVE_PG_PASSWORD=postgres
+EVOLVE_PG_PASSWORD=postgres  # pragma: allowlist secret
 EVOLVE_PG_DBNAME=evolve
 EVOLVE_MODEL_NAME=Azure/gpt-4o
-OPENAI_API_KEY=env://OPENAI_API_KEY
+OPENAI_API_KEY=env://OPENAI_API_KEY  # pragma: allowlist secret
 OPENAI_BASE_URL=env://OPENAI_BASE_URL
 ```
 
@@ -122,7 +118,6 @@ Identify the common cities between my cuga_workspace/cities.txt and cuga_workspa
 | `mode = "auto"` | Registry first, then direct SSE |
 | `mode = "registry"` | Registry only |
 | `mode = "direct"` | Use `url` only |
-| `lite_mode_only = true` | Memory only on the CugaLite path |
 
 ---
 

content/docs/configure/models.mdx

@@ -28,7 +28,7 @@ CUGA supports multiple LLM providers with flexible configuration options. You ca
    ```bash
    # For OpenAI:
    # AGENT_SETTING_CONFIG="settings.openai.toml" #Could be replaced by the settings in ./configurations/models
-   # OPENAI_API_KEY="XXXXXX"
+   # OPENAI_API_KEY="XXXXXX"  # pragma: allowlist secret
    # OPENAI_API_VERSION="2024-08-01-preview"
    ```
 
@@ -58,7 +58,7 @@ MODEL_NAME="gpt-4-turbo"
    # AGENT_SETTING_CONFIG="settings.watsonx.toml" #Could be replaced by the settings in ./configurations/models
    # WATSONX_PROJECT_ID="XXXXXX"
    # WATSONX_URL="https://us-south.ml.cloud.ibm.com"
-   # WATSONX_APIKEY="XXXXXX"
+   # WATSONX_APIKEY="XXXXXX"  # pragma: allowlist secret
    ```
 
 **Default Values:**
@@ -78,7 +78,7 @@ MODEL_NAME="meta-llama/llama-3.1-8b-instruct"
    ```bash
    # For Azure Openai
    # AGENT_SETTING_CONFIG="settings.azure.toml" #Could be replaced by the settings in ./configurations/models
-   # AZURE_OPENAI_API_KEY="XXXXXX"
+   # AZURE_OPENAI_API_KEY="XXXXXX"  # pragma: allowlist secret
    # AZURE_OPENAI_ENDPOINT="XXXXXX"
    ```
 
@@ -96,7 +96,7 @@ MODEL_NAME="gpt-4o"
    ```bash
    # For Groq
    # AGENT_SETTING_CONFIG="settings.groq.toml"
-   # GROQ_API_KEY="XXXX"
+   # GROQ_API_KEY="XXXX"  # pragma: allowlist secret
    ```
 
 **Environment Variable Override:**
@@ -119,7 +119,7 @@ MODEL_NAME="openai/gpt-oss-120b"
 ```bash
    # For OpenRouter
    # AGENT_SETTING_CONFIG="settings.openrouter.toml"
-   # OPENROUTER_API_KEY="XXXX"
+   # OPENROUTER_API_KEY="XXXX"  # pragma: allowlist secret
    # OPENROUTER_BASE_URL="https://openrouter.ai/api/v1"
 ```
 
@@ -146,7 +146,7 @@ RITS (Research Internal Token Service) is an IBM-internal LLM gateway service.
    ```bash
    # For RITS
    AGENT_SETTING_CONFIG="settings.rits.toml"
-   RITS_API_KEY="XXXXXX"
+   RITS_API_KEY="XXXXXX"  # pragma: allowlist secret
    ```
 
 **Default Values:**
@@ -170,7 +170,7 @@ MiniMax-M3 over an OpenAI-compatible API. Default endpoint is global English (`h
 1. Add to your `.env` file:
    ```bash
    AGENT_SETTING_CONFIG="settings.minimax.toml"
-   MINIMAX_API_KEY="XXXXXX"
+   MINIMAX_API_KEY="XXXXXX"  # pragma: allowlist secret
    # Optional: MINIMAX_BASE_URL="https://api.minimax.io/v1"
    ```
 
@@ -192,7 +192,7 @@ MODEL_NAME="MiniMax-M3"
 1. Add to your `.env` file:
    ```bash
    AGENT_SETTING_CONFIG="settings.google.toml"
-   GOOGLE_API_KEY="XXXXXX"
+   GOOGLE_API_KEY="XXXXXX"  # pragma: allowlist secret
    ```
 
 **Environment Variable Override:**
@@ -210,7 +210,7 @@ Ollama is configured as an OpenAI-compatible endpoint. A dummy API key is requir
 2. Add to your `.env` file:
    ```bash
    AGENT_SETTING_CONFIG="settings.ollama.toml"
-   OPENAI_API_KEY="ollama"
+   OPENAI_API_KEY="ollama"  # pragma: allowlist secret
    # Optional: OPENAI_BASE_URL="http://localhost:11434/v1"
    ```
 
@@ -238,7 +238,7 @@ guardrails, and are billed against its entitlement.
 3. Add to your `.env` file:
    ```bash
    AGENT_SETTING_CONFIG="settings.wxo.toml"
-   WXO_API_KEY="XXXXXX"
+   WXO_API_KEY="XXXXXX"  # pragma: allowlist secret
    WXO_INSTANCE_URL="https://api.dl.watson-orchestrate.ibm.com/instances/XXXXXX"
    ```
 
@@ -264,7 +264,7 @@ CUGA supports LiteLLM through the OpenAI configuration by overriding the base UR
    ```bash
    # For LiteLLM:
    # AGENT_SETTING_CONFIG="settings.openai.toml" #Could be replaced by the settings in ./configurations/models
-   # OPENAI_API_KEY="XXXXXX"
+   # OPENAI_API_KEY="XXXXXX"  # pragma: allowlist secret
    # OPENAI_BASE_URL="XXXXXX"
    ```
 
@@ -306,76 +306,43 @@ Use `settings.openai.reasoning-high-32k.toml` if high effort crowds out visible
 
 ### Settings File Content Structure
 
-Each settings file contains configuration for different CUGA agents. You can configure each agent independently with different models and parameters:
+Each settings file configures the same six agent roles: **planner**, **chat**, **final_answer**, **code**, **qa**, and **action**. You can set model and parameters per role.
 
 #### OpenAI Configuration Example
 
 ```toml
-[agent.task_decomposition.model]
-platform = "openai"
-temperature = 0.1
-max_tokens = 1000
-
 [agent.planner.model]
 platform = "openai"
 temperature = 0.1
-max_tokens = 5000
+max_tokens = 16000
 
 [agent.chat.model]
 platform = "openai"
 temperature = 0.1
-max_tokens = 5000
+max_tokens = 16000
 
-[agent.shortlister.model]
+[agent.final_answer.model]
 platform = "openai"
 temperature = 0.1
-max_tokens = 7000
-```
-
-#### WatsonX Configuration Example
+max_tokens = 16000
 
-```toml
-[agent.shortlister.model]
-platform = "watsonx"
-model_name = "meta-llama/llama-4-maverick-17b-128e-instruct-fp8"
+[agent.code.model]
+platform = "openai"
 temperature = 0.1
-max_tokens = 7000
+max_tokens = 16000
 
-[agent.planner.model]
-platform = "watsonx"
-model_name = "meta-llama/llama-4-maverick-17b-128e-instruct-fp8"
+[agent.qa.model]
+platform = "openai"
 temperature = 0.1
-max_tokens = 5000
+max_tokens = 16000
 
-[agent.chat.model]
-platform = "watsonx"
-model_name = "meta-llama/llama-4-maverick-17b-128e-instruct-fp8"
+[agent.action.model]
+platform = "openai"
 temperature = 0.1
-max_tokens = 5000
+max_tokens = 400
 ```
 
-#### OpenRouter Configuration Example
-```toml
-[agent.task_decomposition.model]
-platform = "openrouter"
-temperature = 0.1
-max_tokens = 1000
-
-[agent.planner.model]
-platform = "openrouter"
-temperature = 0.1
-max_tokens = 5000
-
-[agent.chat.model]
-platform = "openrouter"
-temperature = 0.1
-max_tokens = 5000
-
-[agent.shortlister.model]
-platform = "openrouter"
-temperature = 0.1
-max_tokens = 7000
-```
+Other providers (`settings.watsonx.toml`, `settings.azure.toml`, `settings.groq.toml`, `settings.openrouter.toml`, and the rest) use the same six keys. Only `platform` (and optional `model_name`) change.
 
 ## Switching Between Providers